IP Library Granted Patent US 12,676,700
Granted Patent B2
US 12,676,700 · App. 18/002,170 · Granted Jul 7, 2026

Method for transmitting or generating a code using artificial intelligence (AI) / machine learning based on capability information

Inventors: Jongwoong Shin (Seoul, KR); Bonghoe Kim (Seoul, KR); Sungryong Hong (Seoul, KR)
Assignee: LG ELECTRONICS INC.
H04L1/0075H04L1/203H04L41/16H04W8/24H04W72/51
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,676,700
App. No.
18/002,170
Granted
Jul 7, 2026
Kind
B2
Abstract

A method of transmitting and receiving data in a wireless communication system and a device therefor are disclosed. Specifically, the method performed by a user equipment (UE) may comprise transmitting, to a base station, capability information including (i) first information indicating a decoding level of the UE and (ii) second information indicating whether a machine learning is possible, receiving decoding level information from the base station based on the first information, and receiving the downlink data from the base station based on the decoding level information and the second information.

Claims (51)

1 . A method of receiving, by a user equipment (UE), downlink data in a wireless communication system, the method comprising:

transmitting, to a base station, capability information including (i) first information indicating a decoding level of the UE and (ii) second information indicating whether a machine learning is possible;

receiving decoding level information from the base station based on the first information; and

receiving the downlink data from the base station based on the decoding level information and the second information,

wherein receiving the downlink data comprises:

based on the decoding level information indicating a use of a machine learning based code and the second information indicating that the machine learning is possible, generating the machine learning based code based on the machine learning and channel information;

transmitting information on the machine learning based code to the base station; and

receiving the downlink data from the base station based on the machine learning based code.

2 . The method of claim 1 , wherein generating the machine learning based code comprises:

generating n codes based on the machine learning;

generating n pieces of data based on the n codes and the channel information; and

selecting the machine learning based code based on a block error rate (BLER) of the n pieces of data.

3 . The method of claim 1 , wherein receiving the downlink data comprises:

based on the decoding level information indicating the use of a machine learning based code and the second information indicating that the machine learning is not possible, transmitting channel information to the base station; and

receiving information on the machine learning based code from the base station.

4 . The method of claim 1 , wherein receiving the downlink data comprises:

based on the decoding level information indicating a use of a non-machine learning based code, receiving the downlink data from the base station based on the non-machine learning based code.

5 . The method of claim 4 , wherein the non-machine learning based code is a low-density parity check (LDPC) code or a polar code.

6 . The method of claim 1 , further comprising:

receiving a request for the capability information from the base station.

7 . A user equipment (UE) receiving downlink data in a wireless communication system, the UE comprising:

one or more transceivers;

one or more processors operatively connected to the one or more transceivers; and

one or more memories operatively connected to the one or more processors and configured to store instruction performing operations,

wherein the operations comprise:

transmitting, to a base station, capability information including (i) first information indicating a decoding level of the UE and (ii) second information indicating whether a machine learning is possible;

receiving decoding level information from the base station based on the first information; and

receiving the downlink data from the base station based on the decoding level information and the second information,

wherein receiving the downlink data comprises:

based on the decoding level information indicating a use of a machine learning based code and the second information indicating that the machine learning is possible, generating the machine learning based code based on the machine learning and channel information;

transmitting information on the machine learning based code to the base station; and

receiving the downlink data from the base station based on the machine learning based code.

8 . The UE of claim 7 , wherein generating the machine learning based code comprises:

generating n codes based on the machine learning;

generating n pieces of data based on the n codes and the channel information; and

selecting the machine learning based code based on a block error rate (BLER) of the n pieces of data.

9 . The UE of claim 7 , wherein receiving the downlink data comprises:

based on the decoding level information indicating a use of the machine learning based code and the second information indicating that the machine learning is not possible, transmitting channel information to the base station; and

receiving information on the machine learning based code from the base station.

10 . The UE of claim 7 , wherein receiving the downlink data comprises:

based on the decoding level information indicating a use of a non-machine learning based code, receiving the downlink data from the base station based on the non-machine learning based code.

11 . The UE of claim 10 , wherein the non-machine learning based code is a low-density parity check (LDPC) code or a polar code.

12 . The UE of claim 7 , wherein the operations further comprise:

receiving a request for the capability information from the base station.

13 . A method of receiving, by a base station, downlink data in a wireless communication system, the method comprising:

receiving, from a user equipment (UE), capability information including (i) first information indicating a decoding level of the UE and (ii) second information indicating whether a machine learning is possible;

transmitting decoding level information to the UE based on the first information; and

transmitting the downlink data to the UE based on the decoding level information and the second information,

wherein transmitting the downlink data comprises:

based on the decoding level information indicating a use of a machine learning based code and the second information indicating that the machine learning is possible, receiving information on machine learning based code based on the machine learning and channel information from the UE; and

transmitting the downlink data to the UE based on the machine learning based code.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 16, 2022
From: SHIN, JONGWOONG; KIM, BONGHOE; HONG, SUNGRYONG
To: LG ELECTRONICS INC.
Reel/Frame 062129/0983 →
Continuity (1)
Related Publication 20230231653A1 · Jul 20, 2023
References Cited (15)
US 20160029211A1 · Furuta · 2016 [cited by applicant]
US 20180174050A1 · Holt et al. · 2018 [cited by applicant]
US 20190079898A1 · Xiong et al. · 2019 [cited by applicant]
US 20190305882A1 · Wang · 2019 [cited by examiner]
US 20210110261A1 · Lee · 2021 [cited by examiner]
US 20210160149A1 · Ma · 2021 [cited by examiner]
US 20210399770A1 · Xu · 2021 [cited by examiner]
US 20220104122A1 · Maleki · 2022 [cited by examiner]
US 20220360477A1 · Sun · 2022 [cited by examiner]
US 20220393781A1 · Kim · 2022 [cited by examiner]
KR 1020170095275 · 2017 [cited by applicant]
WO 2020035685 · 2020 [cited by applicant]
Korean Intellectual Property Office Application No. 10-2023-7000378, Office Action dated Feb. 18, 2025, 5 pages. [cited by applicant]
PCT International Application No. PCT/KR2020/007899, International Search Report dated Mar. 8, 2021, 4 page. [cited by applicant]
Devamane et al., “Recurrent neural network based turbo decoding algorithms for different code rates,” Journal of King Saud University—Computer and Information Sciences, Apr. 2020, 15 pages. [cited by applicant]